R ile analize başlarken¹

1 R nerede kullanılır

• Veri düzenleme
• İstatistik analiz
• Web sayfası hazırlama (Statik/Dinamik)
• Sunum hazırlama
• Programlama
• Otomatik, periodik ve tekrarlanabilir rapor hazırlama
• pdf, html, ppt oluşturma
• tez yazma
• kitap yazma
• rapor şablonu oluşturma
• …

---

2 R generation

R yıllar içinde çok fazla değişim gösterdi

https://rss.onlinelibrary.wiley.com/doi/10.1111/j.1740-9713.2018.01169.x

:scale 30%

---

3 R yükleme

http://www.youtube.com/watch?v=XcBLEVknqvY

3.1

3.2 R-project

https://cran.r-project.org/

---

3.3 RStudio

---

3.4 RStudio

https://www.rstudio.com/

https://www.rstudio.com/products/rstudio/download/

https://moderndive.com/2-getting-started.html

---

---

3.4.1 RStudio eklentileri

• Discover and install useful RStudio addins

https://cran.r-project.org/web/packages/addinslist/README.html

https://rstudio.github.io/rstudioaddins/

devtools::install_github("rstudio/addinexamples", type = "source")

---

4 MacOS için

4.1 X11

https://www.xquartz.org/

4.2 Java OS

https://support.apple.com/kb/dl1572

---

5 R zor şeyler için kolay, kolay şeyler için zor

• R makes easy things hard, and hard things easy

• Aynı şeyi çok fazla şekilde yapmak mümkün

R Syntax Comparison::CHEAT SHEET

https://www.amelia.mn/Syntax-cheatsheet.pdf

---

#RStats — There are always several ways to do the same thing… nice example on with the identity matrix by @TeaStats https://t.co/O3GXdPiM32

— Colin Fay 🤘 (@_ColinFay) April 1, 2019

---

6 R paketleri

6.1 Neden paketler var

---

I love the #rstats community.
Someone is like, “oh hey peeps, I saw a big need for this mundane but difficult task that I infrequently do, so I created a package that will literally scrape the last bits of peanut butter out of the jar for you. It's called pbplyr.”
What a tribe.

— Frank Elavsky ᴰᵃᵗᵃ ᵂᶦᶻᵃʳᵈ (@Frankly_Data) July 3, 2018

---

---

6.2 Paketleri nereden bulabiliriz

• Available CRAN Packages By Name
  https://cran.r-project.org/web/packages/available_packages_by_name.html

• CRAN Task Views
  https://cran.r-project.org/web/views/

• Bioconductor
  https://www.bioconductor.org

• RecommendR
  http://recommendr.info/

• pkgsearch
  CRAN package search
  https://github.com/metacran/pkgsearch

• CRANsearcher
  https://github.com/RhoInc/CRANsearcher

• Awesome R
  https://awesome-r.com/

---

6.3 Kendi paket evrenini oluştur

• pkgverse: Build a Meta-Package Universe
  https://cran.r-project.org/web/packages/pkgverse/index.html

---

6.4 R paket yükleme

install.packages("tidyverse", dependencies = TRUE)
install.packages("jmv", dependencies = TRUE)
install.packages("questionr", dependencies = TRUE)
install.packages("Rcmdr", dependencies = TRUE)
install.packages("summarytools")

# install.packages("tidyverse", dependencies = TRUE)
# install.packages("jmv", dependencies = TRUE)
# install.packages("questionr", dependencies = TRUE)
# install.packages("Rcmdr", dependencies = TRUE)
# install.packages("summarytools")

---

6.5 Paket çağırma

# require(tidyverse)
# require(jmv)
# require(questionr)
# library(summarytools)
# library(gganimate)

---

7 R için yardım bulma

# ?mean
# ??efetch
# help(merge)
# example(merge)

---

• Vignette

:scale 80%

---

• RDocumentation https://www.rdocumentation.org

• R Package Documentation https://rdrr.io/

• GitHub

• Stackoverflow

https://stackoverflow.com/

• Google uygun anahtar kelime

---

How I use #rstats
h/t @ThePracticalDev pic.twitter.com/erRnTG0Ujr

— Emily Bovee (@ebovee09) August 10, 2018

---

---

• Google’da ararken [R] yazmak da işe yarayabiliyor.

---

• searcher package 📦

---

• Awesome Cheatsheet https://github.com/detailyang/awesome-cheatsheet

http://cran.r-project.org/doc/contrib/Baggott-refcard-v2.pdf

https://www.rstudio.com/resources/cheatsheets/

• Awesome R

https://github.com/qinwf/awesome-R#readme

https://awesome-r.com/

• Twitter

https://twitter.com/hashtag/rstats?src=hash

---

• Use Reproducible Examples When Asking

Got a question to ask on @SlackHQ or post on @github? No time to read the long post on how to use reprex? Here is a 20-second gif for you to format your R codes nicely and for others to reproduce your problem. (An example from a talk given by @JennyBryan) #rstat pic.twitter.com/gpuGXpFIsX

— ZhiYang (@zhiiiyang) October 18, 2018

• Keeping up to date with R news
  https://masalmon.eu/2019/01/25/uptodate/

---

8 R studio ile proje oluşturma

https://support.rstudio.com/hc/en-us/articles/200526207-Using-Projects

---

9 RStudio ile veri yükleme

https://support.rstudio.com/hc/en-us/articles/218611977-Importing-Data-with-RStudio

---

9.1 Excel

9.2 SPSS

9.3 CSV

---

10 Veriyi görüntüleme

Spreadsheet users using #rstats: where's the data?#rstats users using spreadsheets: where's the code?

— Leonard Kiefer (@lenkiefer) July 7, 2018

---

11 Veriyi görüntüleme

# library(nycflights13)
# summary(flights)

View(data)

data

head

tail

glimpse

str

skimr::skim()

---

12 Veriyi değiştirme

12.1 Veriyi kod ile değiştirelim

12.2 Veriyi eklentilerle değiştirme

:scale 50%

---

12.3 RStudio aracılığıyla recode

questionr paketi kullanılacak

:scale 50%

---

https://juba.github.io/questionr/articles/recoding_addins.html

---

---

---

13 Basit tanımlayıcı istatistikler

summary()

mean

median

min

max

sd

table()

---

library(readr)
irisdata <- read_csv("data/iris.csv")

Parsed with column specification:
cols(
  Sepal.Length = col_double(),
  Sepal.Width = col_double(),
  Petal.Length = col_double(),
  Petal.Width = col_double(),
  Species = col_character()
)

jmv::descriptives(
    data = irisdata,
    vars = "Sepal.Length",
    splitBy = "Species",
    freq = TRUE,
    hist = TRUE,
    dens = TRUE,
    bar = TRUE,
    box = TRUE,
    violin = TRUE,
    dot = TRUE,
    mode = TRUE,
    sum = TRUE,
    sd = TRUE,
    variance = TRUE,
    range = TRUE,
    se = TRUE,
    skew = TRUE,
    kurt = TRUE,
    quart = TRUE,
    pcEqGr = TRUE)

 DESCRIPTIVES

 Descriptives                                          
 ───────────────────────────────────────────────────── 
                          Species       Sepal.Length   
 ───────────────────────────────────────────────────── 
   N                      setosa                  50   
                          versicolor              50   
                          virginica               50   
   Missing                setosa                   0   
                          versicolor               0   
                          virginica                0   
   Mean                   setosa                5.01   
                          versicolor            5.94   
                          virginica             6.59   
   Std. error mean        setosa              0.0498   
                          versicolor          0.0730   
                          virginica           0.0899   
   Median                 setosa                5.00   
                          versicolor            5.90   
                          virginica             6.50   
   Mode                   setosa                5.00   
                          versicolor            5.50   
                          virginica             6.30   
   Sum                    setosa                 250   
                          versicolor             297   
                          virginica              329   
   Standard deviation     setosa               0.352   
                          versicolor           0.516   
                          virginica            0.636   
   Variance               setosa               0.124   
                          versicolor           0.266   
                          virginica            0.404   
   Range                  setosa                1.50   
                          versicolor            2.10   
                          virginica             3.00   
   Minimum                setosa                4.30   
                          versicolor            4.90   
                          virginica             4.90   
   Maximum                setosa                5.80   
                          versicolor            7.00   
                          virginica             7.90   
   Skewness               setosa               0.120   
                          versicolor           0.105   
                          virginica            0.118   
   Std. error skewness    setosa               0.337   
                          versicolor           0.337   
                          virginica            0.337   
   Kurtosis               setosa              -0.253   
                          versicolor          -0.533   
                          virginica           0.0329   
   Std. error kurtosis    setosa               0.662   
                          versicolor           0.662   
                          virginica            0.662   
   25th percentile        setosa                4.80   
                          versicolor            5.60   
                          virginica             6.23   
   50th percentile        setosa                5.00   
                          versicolor            5.90   
                          virginica             6.50   
   75th percentile        setosa                5.20   
                          versicolor            6.30   
                          virginica             6.90   
 ───────────────────────────────────────────────────── 

---

# install.packages("scatr")

scatr::scat(
    data = irisdata,
    x = "Sepal.Length",
    y = "Sepal.Width",
    group = "Species",
    marg = "dens",
    line = "linear",
    se = TRUE)

---

13.1 summarytools

https://cran.r-project.org/web/packages/summarytools/vignettes/Introduction.html

library(summarytools)
summarytools::freq(iris$Species, style = "rmarkdown")

13.1.1 Frequencies

13.1.1.1 iris$Species

Type: Factor

	Freq	% Valid	% Valid Cum.	% Total	% Total Cum.
setosa	50	33.33	33.33	33.33	33.33
versicolor	50	33.33	66.67	33.33	66.67
virginica	50	33.33	100.00	33.33	100.00
<NA>	0			0.00	100.00
Total	150	100.00	100.00	100.00	100.00

---

summarytools::freq(iris$Species, report.nas = FALSE, style = "rmarkdown", headings = FALSE)

with(tobacco, print(ctable(smoker, diseased), method = 'render'))

with(tobacco,
     print(ctable(smoker, diseased, prop = 'n', totals = FALSE),
           omit.headings = TRUE, method = "render"))

---

summarytools::descr(iris, style = "rmarkdown")

---

descr(iris,
      stats = c("mean", "sd", "min", "med", "max"),
      transpose = TRUE,
      headings = FALSE,
      style = "rmarkdown")

---

# view(dfSummary(iris))

---

dfSummary(tobacco,
          plain.ascii = FALSE,
          style = "grid")

---

# First save the results

iris_stats_by_species <- by(data = iris,
                            INDICES = iris$Species,
                            FUN = descr, stats = c("mean", "sd", "min", "med", "max"),
                            transpose = TRUE)

# Then use view(), like so:

view(iris_stats_by_species, method = "pander", style = "rmarkdown")

---

# view(iris_stats_by_species)

---

data(tobacco) # tobacco is an example dataframe included in the package
BMI_by_age <- with(tobacco,
                   by(BMI, age.gr, descr,
                      stats = c("mean", "sd", "min", "med", "max")))
view(BMI_by_age, "pander", style = "rmarkdown")

---

BMI_by_age <- with(tobacco,
                   by(BMI, age.gr, descr,  transpose = TRUE,
                      stats = c("mean", "sd", "min", "med", "max")))

view(BMI_by_age, "pander", style = "rmarkdown", omit.headings = TRUE)

---

tobacco_subset <- tobacco[ ,c("gender", "age.gr", "smoker")]
freq_tables <- lapply(tobacco_subset, freq)

# view(freq_tables, footnote = NA, file = 'freq-tables.html')

---

what.is(iris)

---

freq(tobacco$gender, style = 'rmarkdown')

---

print(freq(tobacco$gender), method = 'render')

---

13.2 skimr

library(skimr)
skim(df)

---

13.3 DataExplorer

library(DataExplorer)
DataExplorer::create_report(df)

---

13.4 inspectdf

https://github.com/alastairrushworth/inspectdf

---

13.5 Grafikler

# library(ggplot2)
# library(mosaic)
# mPlot(irisdata)

---

ctable(tobacco$gender, tobacco$smoker, style = 'rmarkdown')

---

print(ctable(tobacco$gender, tobacco$smoker), method = 'render')

---

descr(tobacco, style = 'rmarkdown')

print(descr(tobacco), method = 'render', table.classes = 'st-small')

dfSummary(tobacco, style = 'grid', plain.ascii = FALSE)

print(dfSummary(tobacco, graph.magnif = 0.75), method = 'render')

---

Here, building up a #ggplot2 as slowly as possible, #rstats. Incremental adjustments. #rstatsteachingideas pic.twitter.com/nUulQl8bPh

— Gina Reynolds (@EvaMaeRey) August 13, 2018

---

Dreaming of a fancy #Rstats #ggplot #dataviz but still scared of typing #code? @_pvictorr esquisse package has you covered https://t.co/1vIDXcVAAF pic.twitter.com/RlTkptnrNv

— Radoslaw Panczak (@RPanczak) October 2, 2018

---

14 Bazı arayüzler

14.1 Rcmdr

library(Rcmdr)

Rcmdr::Commander()

• A Comparative Review of the R Commander GUI for R

http://r4stats.com/articles/software-reviews/r-commander/

---

14.2 jamovi

https://www.jamovi.org/

https://blog.jamovi.org/2018/07/30/rj.html

---

15 R nereden öğrenilir

https://sbalci.github.io/MyRCodesForDataAnalysis/WhereToLearnR.nb.html

---

16 Sonraki Konular

• RStudio ile GitHub kullanımı
• R Markdown ve R Notebook ile tekrarlanabilir rapor
• Hipotez testleri

---

17 Geri Bildirim

• Geri bildirim için tıklayınız: Geri bildirim formu

---

---

# Save Final Data

saved data after analysis to `Data-After-Analysis.xlsx`.

saveRDS(mydata, "Data-After-Analysis.rds")

writexl::write_xlsx(mydata, "Data-After-Analysis.xlsx")

file.info("Data-After-Analysis.xlsx")$ctime

---

18 Libraries Used

citation()

citation("tidyverse")
citation("foreign")
citation("tidylog")
citation("janitor")
citation("jmv")
citation("tangram")
citation("finalfit")
citation("summarytools")
citation("ggstatplot")
citation("readxl")

---

report::cite_packages(session = sessionInfo())

---

sessionInfo()

---

19 Notes

Completed on 2019-09-24 10:52:42.

Serdar Balci, MD, Pathologist
drserdarbalci@gmail.com
https://rpubs.com/sbalci/CV
https://sbalci.github.io/
https://github.com/sbalci

---

1. Bu bir derlemedir, mümkün mertebe alıntılara referans vermeye çalıştım.↩︎